How many have died in the heatwave?

18 July 2013

The Times (£) broke the news today of how many of us are supposedly dying because of the heatwave: between 540 and 760 people in England in the first nine days.

The figures come from academic research commissioned by the Times, provided by the London School of Hygiene and Tropical Medicine.

Numbers like these pop into the headlines with little provenance or plausibility so often that arguably even the value of academic blessing has melted. One media organisation contacted Full Fact for our take on these figures.

The basics

First, the author Prof. Armstrong is an epidemiological statistician with a history of published research on the effects of hot weather on health, so we're dealing with a specialist.

Secondly, better than just having a job title, the full calculations are available online for anyone to check for themselves.

Thirdly, Prof. Armstrong answered our questions and sent us a copy of the underlying journal article - not the kind of cooperation we've always been able to get hold of.

These are the kinds of numbers factcheckers dream of: traceable, checkable and helpfully explained. It's a shame the Times didn't take the chance to link through to the details.

But more than that, it's actually quite difficult to work out from the article whether the central estimate of 650 excess deaths is actually that much more than we'd expect for this time of year anyway. So:

Is 650 excess deaths over nine days a lot?

466,779 people died in England last year. If we ignore seasonal variation, that's about 11,500 every nine days. So what Prof. Armstrong's research shows is a rate of deaths 4.7—6.6% higher than the crude average for the year.

He explains that:

"The excess is likely to have been overwhelmingly among the elderly, especially those over 75, some of which may have been among people who would have died just a few weeks later if there had been no heatwave."

With lots of people jumping in water, climbing on motorbikes and so on, there are lots of other reasons to think these figures sound about right.

Underneath them is a model of the effect of temperature on health, fully explained in the Journal of Epidemiology and Community Health (£), and based on thirteen years of region-specific data from between 1993 and 2006.

The published figures don't allow you to reproduce the calculations completely from scratch. However, we were able to estimate that the model suggests the increase in risk caused by the first nine days of the heatwave in England risk was roughly 5.9%. That's within the range suggested by a crude comparison of the headline figure to average annual deaths.

650 excess deaths are undeniably important. Nevertheless, the headline numbers look a little less dramatic when you set them next to around 11,000 deaths that might be expected in the same period even without the heatwave.

Full Fact fights bad information

Bad information ruins lives. It promotes hate, damages people’s health, and hurts democracy. You deserve better.